Understanding Uncertainty ( In Polls and Elections)

[bluewater]

Understanding Uncertainty

Two professors work to better visualize probability for the next election.

Across the political spectrum, surprise was a common reaction to the 2016 presidential election. For most people, the predictions leading up to Nov. 8 made it seem as if Hillary Clinton was all but certain to become the first female president. Depending on your perspective, waking up on Nov. 9 felt like an icy splash of water or a thrilling upset.

And some of that surprise, say Northwestern professors Jessica Hullman and Steve Franconeri, can be credited to the predictive charts and graphs used in 2016. Hullman, who holds appointments in both the McCormick School of Engineering and the Medill School of Journalism, Media, Integrated Marketing Communications, studies how data visualizations can help people understand uncertainty. Franconeri, a psychology professor in the Weinberg College of Arts and Sciences and director of the Northwestern Cognitive Science Program, studies how people process visual information.

These two researchers say incomplete or oversimplified visuals, paired with our human tendency to see what we want to see, contributed to the difficulty many people had in fully grasping the range of possible election outcomes.

Here, journalism, computer science and psychology collide as Hullman and Franconeri break down some of what went wrong in 2016 visualizations, and how data journalists can show probabilities more effectively in 2020 — and beyond — to better prepare voters.

https://magazine.northwestern.edu/exclusives/understanding-uncertainty/?linkId=81365723

Well worth looking at the linked article for the analysis and graphics.